The nature of the specific genetic factors contributing to UC and CD is poorly defined. Given the immunopathogenetic features of IBD, the major histocompatibility complex (MHC) is likely contain one or more of the susceptibility genes for UC or CD on their subsets. In support of this concept, we have obtained data which implicates the involvement of the HLA class II genes in the development of IBD and suggests that this susceptibility is likely to be heterogeneous. These data include HLA associations, HLA transmission distortion, the distribution of the UC specific anti-neutrophil cytoplasmic antibodies (ANCAs) within families, and HLA associations with ANCA. Intriguingly, we have recently observed an even greater association between the tumor necrosis factor (TNF) locus (located in the MHC region) and CD. This suggests either that TNF may play an independent and possibly even more important role than the HLA class II genes in CD susceptibility, or that other genes on the short arm of chromosome 6 provide the primary susceptibility for CD and the associations of both HLA class II genes and TNF genes are secondary to the linkage disequilibrium with these genes. Thus, we propose to intensively investigate a number of candidate genes and gene regions in the MHC and its flanking regions to localize these primary susceptibility gene(s) and their alleles in the MHC and/or adjacent region for each of the different forms of UC and CD with both linkage and association approaches. To accomplish this goal we will (1) subclassify and/or group the disease(s) into more genetically homogeneous populations by characterizing the clinical and subclinical features in UC and CD patients and their relatives and examining familiarity of the various clinical and subclinical traits and their relationships to genetic markers in the HLA class I, II and III regions; (2) perform linkage analysis primarily by nonparametric linkage approaches to identify gene regions that link with each form of UC and of CD; and (3) identify the primary susceptibility gene(s) or haplotype(s) for each form of UC and of CD in the MHC region by utilizing an association approach with closer markers in the linked region in our case-control panel and families (i.e., haplotype relative risk approach) and a trans-ethnic gene mapping approach by expanding our Caucasian case-control panel to other ethnic groups. Success in this effort will increase our ability to delineate the genetic and etiologic heterogeneity of IBD and identify which allele/haplotypes provide the highest risk, and thus provide the basis for the eventual molecular delineation of the specific genes and their variations that predispose to UC and CD in this gene region.